Quadrature Response Spectra Deep Neural Based Behavioral Pattern Analytics for Epileptic Seizure Identification

Abstract

Brain Electroencephalogram (EEG) signals comprise of essential information of the brain and are extensively utilized in assisting analysis of epilepsy.  By analyzing brain behavioral pattern, accurate classification of different epileptic states can be made. Behavioral pattern analysis applying EEG signals is getting as well as further awareness over the past few years. Because, EEG signals were boisterous and non-linear, it has a demanding mission for designing clever method to offer maximum accurate classification of different epileptic states. In this work a method called, Quadrature Response Spectra-based Gaussian Kullback Deep Neural (QRS-GKDN) Behavioral Pattern Analytics for epileptic seizure is introduced. QRS-GKDN is dividing three processes. They are preprocessing the EEG signals using Quadrature Mirror Filter, applying Power Frequency Spectral and Response Spectra-based Feature Extraction for Behavioral Pattern Analytics. First, by eradicating intrusion as well as noise produced from input EEG signals, Quadrature Mirror Filter function is employed. Second with the processed EEG signals as input, relevant feature extractions are made for behavioral pattern analytics using Power Frequency Spectral and Response Spectra function. With the aid of this behavioral pattern analytics classification of different epileptic states can be made. Finally, with the extracted features Gaussian Kullback–Leibler Deep Neural Classification for Epileptic seizure identification. In addition, proposed method is analyzed as well as contrasted by dissimilar samples. Proposed method of outcome has superior forecast in a computationally efficient manner for identifying epileptic seizure based on the analyzed behavioral patterns with less error and validation time.